Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add kumaran-is/claude-code-onboarding --skill ai-playbookgit clone --depth 1 https://github.com/kumaran-is/claude-code-onboardingWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/kumaran-is/claude-code-onboarding/ai-playbook)<a href="https://agentmods.dev/skills/kumaran-is/claude-code-onboarding/ai-playbook"><img src="https://agentmods.dev/badge/skills/kumaran-is/claude-code-onboarding/ai-playbook/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/kumaran-is/claude-code-onboarding/ai-playbook"><img src="https://agentmods.dev/badge/skills/kumaran-is/claude-code-onboarding/ai-playbook.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00182 | $0.01825 |
| Opus 5 | $0.00091 | $0.00912 |
| Sonnet 5 | $0.00036 | $0.00365 |
| Haiku 4.5 | $0.00018 | $0.00183 |
Grade A, and why
ai-playbook scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 131 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI Application Playbook
Iron Law: Start one tier below where you think you need to be. Most "agent" problems are workflows; most "workflow" problems are a single model call. Use the smallest pattern that works.
This skill encodes a production-grade framework for deciding when to use AI and how to ship it safely. The full playbook is in playbook.md (10k+ words across four layers); load it only when needed using the routing below.
Prime Directive
Use AI for translation between the messy world and your clean system — never as the clean system itself.
Everything the model reads is untrusted input. Everything the model writes is an untrusted proposal until validated.
Fast triage (use this first)
Before any AI design work, ask in order:
- Is the input messy / unstructured / ambiguous? If no → deterministic code, stop.
- Can a downstream verifier (human or code) catch errors? If no → don't use AI for the decision.
- Is this on the critical latency path with no fallback? If yes → don't use AI here.
- Pick the smallest pattern that works: single model call in a workflow → ReAct loop in a state machine → planning agent → autonomous agent. Start at #1.
If the answer to all three is "AI is appropriate," route to the relevant layer below.
When to load the full playbook
Read playbook.md and apply the matching layer based on what the user is doing:
| User is doing... | Load these layers from playbook.md |
|---|---|
| Deciding whether to add AI to a feature | Layer 1 (Principles) |
| Architecting an AI feature | Layer 1 + Layer 2 (Operating Controls) |
| Writing code: schema validation, output contract, retries, fallback | Layer 2 §2.4, §2.5; Layer 3 §3.1, §3.5 |
| Writing prompts or designing prompts for security | Layer 2 §2.5; Layer 3 §3.2 (adversarial test catalog) |
| Building or reviewing an agent | Layer 2 §2.5, §2.6; Layer 3 §3.2 |
| Designing or evaluating RAG | Layer 2 §2.7; Layer 3 §3.2 retrieval-poisoning row |
| Picking a model / deciding to fine-tune | Layer 2 §2.9 |
| Sizing cost and latency | Layer 2 §2.10, §2.11; Layer 3 §3.4 |
| Designing eval set / regression tests | Layer 2 §2.14; Layer 3 §3.2 |
| Preparing for launch | Layer 4 §4.1 (unified launch checklist) |
| Setting up rollout or canary | Layer 4 §4.2 |
| Handling an incident | Layer 4 §4.3 |
| Testing kill switch | Layer 4 §4.4 |
| Onboarding to AI workflow (first 30 days) | Layer 4 §4.5 |
| Avoiding overbuilding | Layer 4 §4.6 |
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 6d ago First seen · 131 lines · 0 tokens per session scan A ec4068c7c068
ai-playbook is a skill published in the GitHub repository kumaran-is/claude-code-onboarding (35 stars, last pushed 2mo ago), licensed MIT. It adds 182 tokens to every session and 1,825 once invoked, about $0.0009 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
dspy
DSPy: declarative LM programs, auto-optimize prompts, RAG.
ai-engineering-toolkit
6 production-ready AI engineering workflows: prompt evaluation (8-dimension scoring), context budget planning, RAG pipeline design, agent security audit (65-point checklist), eval harness building, and product sense coaching.
llm-app-patterns
Production-ready patterns for building LLM applications. Covers RAG pipelines, agent architectures, prompt IDEs, and LLMOps monitoring. Use when designing AI applications, implementing RAG, building agents, or setting up LLM observability.
llm-security
Use for authorized security assessment of LLM applications and AI agents, including prompt injection, tool abuse, RAG exposure, memory poisoning, and model supply-chain risks.
ai-product
Every product will be AI-powered. The question is whether you'll build it right or ship a demo that falls apart in production.
llm-ops
LLM Operations -- RAG, embeddings, vector databases, fine-tuning, prompt engineering avancado, custos de LLM, evals de qualidade e arquiteturas de IA para producao.